Chaewoon Ki
Undergraduate student @ KAIST CS
I am an undergraduate student in the School of Computing at KAIST, currently working as an undergraduate researcher at the MLAI Lab in KAIST Graduate School of AI.
My long-term goal is to build AI systems that go beyond surface-level language processing: systems that can genuinely understand, retain, and use context in meaningful ways.
For more, see the Projects section.
Research
My research aims to make large language models more reliable and adaptive when they reason over information that changes across time and context. I focus on giving models a sense of memory: the ability to retain what matters, recall it at the right moment, and adjust their behavior as the environment evolves.
My current research focuses include:
- Memory for LLM Agents: Designing memory architectures that let agents accumulate and reuse context over long horizons, separating transient facts from durable behavioral patterns.
- Context-Aware Reasoning: Improving how models ground their reasoning in the right information at the right time, rather than treating every query in isolation.
- Personalized Agents: Developing methods that let agents adapt to individual users and goals, so interactions become more relevant the longer they continue.
- Practical AI Systems: Building systems that hold up in real-world use, not only on benchmarks, with attention to reliability and efficiency.
Education
Korea Advanced Institute of Science and Technology (KAIST)
B.S. in Computer Science
Hansung Science High School
Experience
DAVIAN Lab, KAIST Graduate School of AI
Undergraduate Researcher
Worked on LLM-based personalized web agents and contributing to a co-authored research paper.
SynBi Lab, KAIST Bio and Brain Engineering
Undergraduate Researcher
Participated in research on drug candidate discovery and protein activity prediction.
Research Intern
Worked on battery SOH prediction, conducted comparative ML/DL experiments, and contributed to a research paper.